Confidently Incorrect: Misunderstanding “How Much Is Too Much” for Fair Use

Check out our series of posts marking the 50th anniversary of the Copyright Act of 1976, as well as a variety of other Fair Use Week blogs and events around the country.

Of the four statutory fair use factors, the “the amount and substantiality of the portion used in relation to the copyrighted work as a whole,” may be the one that generates the most confident wrong answers. In my experience, when I ask people whether a use is fair, many will reach for a number: no more than 10%, no more than 1,000 words, or no more than one chapter. These thresholds feel authoritative, and they are stated with conviction. Yet they are legally meaningless, as Kenny Crews documents in his 2001 article The Law of Fair Use and the Illusion of Fair-Use Guidelines—a case study in confident incorrectness.

The story of how this happened is worth a brief detour. The most familiar thresholds—“10% of a work,” and “no more than 1,000 words”—derive from the Classroom Guidelines, from page 68 of the House Report, a negotiated agreement between publishers and educator representatives developed alongside the 1976 Copyright Act and included in House committee reports. That proximity to Congress gave them the appearance of official sanction. 

As Crews explains, the Guidelines were never enacted as law, they bear no systematic relationship to the four-factor analysis Congress actually put in the statute, and courts that have cited the guidelines have generally conducted their fair use analysis without actually relying on them, treating them as secondary context at best. The problem Crews identifies goes further than mere inaccuracy: guidelines presented as authoritative tend to suppress uses that would be fully defensible under the law, while providing false assurance about others. By replacing the four-factor analysis with a simple rule or number,  they’ve made educators and librarians less capable of reasoning about fair use, not more.

This is especially unfortunate for the third factor because, freed from the guidelines framework, it can be one of the most practically useful questions in the whole analysis.

What Campbell Established

Before Campbell v. Acuff-Rose Music (1994), the Supreme Court had limited guidance on how to actually apply the third factor. In Sony v. Universal Studios (1984), the Court established that copying an entire work would ordinarily weigh against fair use, though without saying what exactly the exception to that “ordinary” circumstance consists of.  Harper & Row v. Nation (1985) added the important clarification that quality matters as well as quantity: a quantitatively small taking can still weigh heavily against fair use if it captures the “heart” of the work. But neither case offered a general principle that courts could apply across situations.

Campbell filled that gap by instructing that the third factor calls for examining whether the amount taken was “reasonable in relation to the purpose of the copying.” That explanation is the analytical move that makes the factor genuinely useful. It also renders the Classroom Guidelines obsolete for third-factor analysis: once the question is “was this amount reasonable given your purpose,” a percentage threshold is simply the wrong answer to the right question. Anthony Reese’s 2015 empirical study of 61 post-Campbell appellate fair use decisions,How Much Is Too Much?: Campbell and the Third Fair Use Factor, documents how thoroughly the circuits absorbed this framework. Among other things, Reese finds that this “reasonable” standard has a significant impact, so much so that uses of the entire work not only come out in favor of fair use, but courts often also find that the third factor itself resolves in that direction as well. 

Why Entire-Work Copying Can Favor Fair Use

This last point runs counter to what the percentage-based guidelines reinforce. Numerous courts have found that copying an entire work not only doesn’t preclude fair use, but can actually weigh in favor of it under the third factor. Reese traces this most clearly in cases involving new technological uses.  Thumbnails for image searches are the central example, such as Kelly v. Arriba Soft, where significant portions of an image was found reasonable because the purpose was not substitution for the original. The third factor in Kelly weighed in favor of fair use precisely because the complete copy was what the purpose required, and because that purpose served a different market function than the original work.

This is why the third factor functions as a useful substitution heuristic, not just a quantity check. The first factor’s inquiry into the “purpose and character of the use” addresses substitution conceptually, but it can be abstract. The third factor makes it concrete: the question is not so much, “does this use threaten the market?,” but, “have I taken only what my purpose actually required—and does what I’ve taken risk standing in for the original?” As Michael Donaldson points out in his article Refuge from the Storm: A Fair Use Safeharbor for Nonfiction Works, the practical rule is to limit your use to no more than you actually need. When applied honestly, that principle both disciplines excessive copying and clarifies—like the thumbnail cases—where copying the whole thing is genuinely what the purpose requires.

Non-Transformative Uses and the Georgia State Case

The reasonableness framework works clearly when the use is highly transformative. Though it becomes considerably harder when the use isn’t. Particularly so with multiple copying for classroom use, which has been the center of the most contested educational copyright litigation.

Cambridge University Press v. Patton (11th Cir. 2014), the Georgia State e-reserves case, is the most important recent judicial treatment of this dynamic. The district court had applied a bright-line rule: excerpts within a 10%-or-one-chapter baseline automatically favored fair use under the third factor. This was the Classroom Guidelines reasoning translated directly into judicial analysis—precisely what Crews warned against. The Eleventh Circuit rejected it, holding that courts cannot escape a reasoned analysis of the amount used.

What followed on remand was a painstaking, excerpt-by-excerpt review of dozens of individual works by the District Court—not a numerical formula applied across the board, but a genuine assessment of each claimed use on its own facts.  Taking a closer look, the District Court found on remand that most digital excerpts downloaded for free by Georgia State University students were in fact protected under fair use.

The Eleventh Circuit’s reasoning on the broader stakes connects the third factor directly to copyright’s incentive structure. As the court explained, “[t]o further the purpose of copyright, we must provide for some fair use taking of copyrighted material. But if we set this transaction cost too high by allowing too much taking, we run the risk of eliminating the economic incentive for the creation of original works that is at the core of copyright and—by driving creators out of the market—killing the proverbial goose that laid the golden egg.” (I quibble with the court about the actual incentives at play here; for most of the authors in the works at issue, the incentive likely was not primarily monetary.)

Practical Takeaways

I’ve advised many, many people on questions of fair use, and I have to admit I often ask as my second or third question “how much did you use”? And then, armed with an answer, usually in the form of “30 seconds” or “two chapters” (typically with some explanation of the purpose of the use), I begin to form an opinion about fair use.

That’s not a contradiction of everything I’ve said above. It’s actually the reasonableness standard at work. When someone tells me “two chapters” and explains why they needed them, I’m already doing the Campbell analysis: I’m asking whether that amount makes sense given what they’re trying to accomplish. The number isn’t the answer; it’s the prompt that gets the analysis started. Think of it as a gut check that either clears the runway for the rest of the fair use analysis or flags early that there might be a problem worth examining more carefully. If someone tells me they used 80% of a book to make a point they could have made with a paragraph, that’s already illuminating before we go any deeper into factor one. The third factor is, in this sense, a first filter; it’s imprecise, intuitive, and for that reason, often the most revealing question you can ask.